The present invention relates to an image processing method and device for representing data of respective dots on a screen by image data in a plurality of polygons.
In video game devices, etc., it is necessary to dynamically display a number of objects, such as characters, etc., on backgrounds on video screens. Each object is represented by a polyhedron. Each polygonal element constituting the object is called Polygon. A polygon is defined as a closed plane figure bounded by straight-line segments (sides) intersecting a points called vertices. Polygons of 3, 4, 5, 6, 7, 8, 10 and 12 sides are called, respectively, triangles, quadrilaterals, pentagons, hexagons, heptagons, octagons, decagons, and dodecagons. Video screens are constituted by superposing image data in a plurality of polygons on backgrounds.
A video screen is constituted by a number of dots. Image data in Polygons are written in respective dots in an area for an object to be displayed. Each Polygon is constituted by coordinate values, and image data, such as color data, etc., of each summit of a polygonal plane of the Polygon. Coordinate values of each summit define an area to be written in, and image data, such as color data, etc., are written in dots of the area. Each object is displayed by combinations of a plurality of Polygons.
But because a video screen is constituted by dots, when superposed errors due to quantization, such as location computation, etc., empty portions between Polygons which are intended to connected with each other occur. For example in FIG. 1A, a one-dot gap due to errors in location computation, etc. is present between Polygons X and Y to be correctly connected with each other; and when a video screen is horizontally scanned, as shown in FIG. 1B, empty dots (holes) without image data is present between the Polygons X and Y.
To prevent occurrences of such empty dots between Polygons, conventionally respective Polygons are enlarged by one dot; usually Polygons are superposed on each other at adjacent portions thereof, so that no empty dots take place at worst at the connection therebetween.
But it has been a problem that conventionally all Polygons are enlarged, the processing for which needs an extra time. Another problem is that enlarging one dot adversely deforms a shape of all Polygons. Further another problem is all Polygons are enlarged, with the result that an extra time is needed to transmit image data in Polygons.